Protocol for spatiotemporal analysis of climate-induced negative sentiment in social media

Climate-induced stressors are associated with population-level mental well-being, yet reproducible workflows linking environmental data to social media sentiment remain scarce. Here, we present a protocol for analyzing spatiotemporal associations between climate-induced stressors and negative sentiment from geotagged social media data. We describe steps for preprocessing large-scale geotagged X data, performing multilingual sentiment analysis with Linguistic Inquiry and Word Count (LIWC-22), integrating climate and health indicators, and fitting a Bayesian spatiotemporal Poisson model. This protocol supports analyses of associations across diverse climate stressors and European regions. For complete details on the use and execution of this protocol, please refer to Al-Ahdal et al. 1 and Al-Ahdal et al. 2